summaryrefslogtreecommitdiffstats
path: root/ldap/servers
Commit message (Collapse)AuthorAgeFilesLines
* Resolves: #210947Noriko Hosoi2006-10-2525-39/+39
| | | | | Summary: parameterizing the hardcoded paths (phase 3. installed binaries, change log, setup) Comment #23
* Related: 210736Nathan Kinder2006-10-171-1/+1
| | | | Phase one of implementing new GNU Automake/Autoconf build system.
* enable pam_passthru for fedora ds buildsRich Megginson2006-10-171-5/+6
|
* Resolves: #204808Noriko Hosoi2006-10-141-1/+1
| | | | | spurious search timeouts (comment #7) fixed anoter int vs. time_t type mismatch.
* Bug(s) fixed: 204623Rich Megginson2006-10-131-2/+2
| | | | | | | | | | | | | Bug Description: deleting attributes when changing password causes server crash Reviewed by: nhosoi (Thanks!) Fix Description: From Michal: "The function mod2smod does not check for mod->mod_bvalues being NULL and tries to dereference it (modutil.c:370). This function happens to be called only by slapi_mods_get_{first,next}_smod(), which are in turn called only by check_trivial_words() in pw.c; this is why the crash appears only when checking password syntax." I added the same check for the mod_values case - even though the code says this should never be called, better to be safe than sorry. Platforms tested: RHEL4
* [208672] parameterizing the hardcoded paths (phase 2. db, log, lock, pid, ↵Noriko Hosoi2006-10-1317-750/+720
| | | | | | | | | | | | cert, and instance dir) 1) introduced these 3 new config attributes: nsslapd-lockdir, nsslapd-tmpdir, nsslapd-certdir, and eliminated: nsslapd-instancedir. 2) replaced the hardcoded paths with the corresponding attribute value in the server as well as in the create/remove instance codes. 3) moved snmp stats file to the nsslapd-tmpdir 4) moved the server instance dir to <prefix>/lib/<brand>-ds/slapd-<id>
* Bug(s) fixed: 185780Rich Megginson2006-10-121-0/+1
| | | | | | | | | | | Bug Description: one byte memory leak in modify Reviewed by: nhosoi (Thanks!) Files: see diff Branch: HEAD Fix Description: Just call slapi_ch_free_string() with the mod->mod_type. This is safe to call with NULL. Platforms tested: RHEL4 Flag Day: no Doc impact: no
* Bug(s) fixed: 183903Rich Megginson2006-10-121-0/+1
| | | | | | | | | Bug Description: Memory leak in ldbm_config.c:replace_ldbm_config_value Reviewed by: nhosoi (Thanks!) Fix Description: Just needed to call slapi_mods_done(&smods) after the call to slapi_modify_internal_pb(). This is the same as in the other places in the server that perform an internal modify operation. Platforms tested: RHEL4 Flag Day: no Doc impact: no
* Bug(s) fixed: 210120Rich Megginson2006-10-101-0/+8
| | | | | | | | | Bug Description: PAM passthru plugin causes directory server to crash Reviewed by: nhosoi (Thanks!) Fix Description: If the DN given in the BIND request is bogus i.e. not a valid DN (at least not one that ldap_explode_dn can parse), we should just skip the PAM processing and just report a reasonable error to the client. Similarly, if the map method says to lookup the pam ID from the bind DN entry, and the entry cannot be found, just report an error and skip pam processing. Platforms tested: FC5 Flag Day: no Doc impact: no
* Bug(s) fixed: 210075Rich Megginson2006-10-098-8/+8
| | | | | | | | | Bug Description: Change version to 1.0.3 Reviewed by: nkinder, nhosoi (Thanks!) Fix Description: Update version strings and etc. to 1.0.3 Platforms tested: FC5 Flag Day: no Doc impact: no
* Bug(s) fixed: 199321Rich Megginson2006-10-091-1/+4
| | | | | | | | | Bug Description: incorrect base64 encoding of SHA passwords crashes server Reviewed by: nkinder (Thanks!) Fix Description: Check the return value of ldif_base64_decode to see if it is less than zero first before the other comparisons. This is the error condition, so we can just return an error. Additionally, the other comparisons with the unsigned should be ok since we know that the hash_len is a positive number. Platforms tested: FC5 Flag Day: no Doc impact: no
* [204808] spurious search timeoutsNoriko Hosoi2006-10-053-4/+7
| | | | Fixed a problem that ldapsearch reports random timeouts on 64-bit machine.
* [207427] parameterizing the hardcoded paths (Comment #15)Noriko Hosoi2006-09-2711-273/+392
| | | | phase 1. parameterizing config, schema and ldif directory
* 208058 - Copy decrypted SASL data starting from buffer offset instead of ↵Nathan Kinder2006-09-261-1/+2
| | | | always copying from the beginning of the buffer.
* [206724] Processed: Replacing PR_SetNetAddr with PRLDAP_SET_PORT for IPv6 ↵Noriko Hosoi2006-09-241-1/+1
| | | | | | support comment#5: Fixed a stupid copy and paste bug...
* [206724] Replacing PR_SetNetAddr with PRLDAP_SET_PORT for IPv6 supportNoriko Hosoi2006-09-153-163/+164
| | | | | | | | slapi-private.h: introduced PRLDAP_SET_PORT to set port to the port field in PRNetAddr. A copy of the same macro in LDAP C SDK (v6). Note: once NSPR provides an equivalent API, we may want to replace this macro with the one. (the NSPR compatibility issue remains, though.) connection.c, daemon.c: replaced PR_SetNetAddr with PRLDAP_SET_PORT.
* Bug(s) fixed: 206662Rich Megginson2006-09-153-19/+24
| | | | | | | | | | | | | Bug Description: PAM passthru: ENTRY map method not working and schema incorrect Reviewed by: nhosoi (Thanks!) Fix Description: 1) Rename all occurrences of pamMapMethod to pamIDMapMethod 2) The parsing code for the map method was just plain wrong - it wasn't incrementing the pointer correctly. 3) This code: if (one == two == three == PAMPT_MAP_METHOD_NONE) - is not correct. Platforms tested: RHEL4 Flag Day: no Doc impact: no
* minor change to test commit email notificationRich Megginson2006-09-151-2/+1
|
* Bug: 205456Rich Megginson2006-09-0619-23/+21
| | | | | | Description: Enable DS Core build/install on HP-UX Fix Description: remove some bogus adminutil and nsdshttpd stuff from the plugin makefiles Reviewed by: nhosoi (Thanks!)
* Bug(s) fixed: 205456Rich Megginson2006-09-061-3/+0
| | | | | | | | | | | | | | | | | | | | | | | Bug Description: Enable core DS build Reviewed by: nkinder, nhosoi (Thanks!) Fix Description: I turned off all of the other components in the build, such as adminserver, setuputil, clients, etc. and enabled the packaging step when building just the core DS. Now, when you build just the core DS, you get a slapd.tar.gz which you unpack in the server root you create e.g. mkdir /opt/rhds ; cd /opt/rhds ; tar xfz slapd.tar.gz In order to create an instance, you have to use the ds_newinst.pl script as described here - http://directory.fedora.redhat.com/wiki/Install_Guide#Installing_just_the_core_directory_server I also got rid of several references to adminutil that are not needed anymore. Platforms tested: RHEL4 64 Flag Day: Yes. In order to build the full setuputil/adminserver package, you must specify USE_SETUPUTIL=1 USE_ADMINSERVER=1 etc. on the make/gmake command line. Doc impact: no QA impact: should be covered by regular nightly and manual testing New Tests integrated into TET: I need to change the acceptance tests to be able to test just the core DS package.
* 204517 - Cleaned up usage of ber_printf to use new ber typesNathan Kinder2006-09-0114-35/+33
|
* [204566] Adjustment for new LDAP C SDK (ber_get_next_buffer_ext)Noriko Hosoi2006-08-311-29/+47
| | | | | | | | | Changes: 1) When ber_get_next_buffer_ext fails, not just LBER_DEFAULT but LBER_OVERFLOW can be set to the ber tag. Thus, LBER_OVERFLOW is also checked in the error check now. 2) The too large length error used be checked as "errno == EMSGSIZE", which is replaced with "tag == LBER_OVERFLOW"
* 204517 - Use new ber types throughout the server code.Nathan Kinder2006-08-3128-240/+240
|
* [204563] Solaris build problemNoriko Hosoi2006-08-302-8/+14
| | | | removed "-z defs" from the command line building libslapd.so and libback-ldbm.so.
* 204410 - Pick up new ldapcsdk, nspr, and sasl componentsNathan Kinder2006-08-283-7/+8
|
* 204355 - Remove winsync code from ldapserver treeNathan Kinder2006-08-2821-7640/+0
|
* [203214] RHDS fails to start on HP-UX 11.23.Noriko Hosoi2006-08-241-6/+14
| | | | Fix Descrition: Stopped using basename and dirname.
* 203043 - Optimizations in password generationNathan Kinder2006-08-211-21/+13
|
* 203043 - Support password generation when using the password modify extended ↵Nathan Kinder2006-08-181-42/+122
| | | | operation
* 202872 - Allow password modify extended operation when using SASL privacy layerNathan Kinder2006-08-176-4/+30
|
* use slapi_entry_attr_get_bool instead of slapi_entry_attr_get_int for the ↵Rich Megginson2006-05-251-4/+4
| | | | fallback and secure config attrs
* [192901] DS returns an entry that does not match the filterNoriko Hosoi2006-05-241-1/+1
| | | | If NOTFOUND is returned from new idl, keep ALLID to check with the filter later.
* [191832] changing password sometimes crashes the server [Admin Server ↵Noriko Hosoi2006-05-191-2/+2
| | | | | | password always remembers initial password on (part 2)] acl_access_allowed: changed to check the given entry address with the one in aclpb. If they don't match, even if the sdn's match, let acl re-evaluate.
* 190724 - Array initialization needed to be changed to fix a HP-UX PA ↵Nathan Kinder2006-05-051-1/+3
| | | | compilation error
* 190724 - Evaluate ACIs before checking password syntaxNathan Kinder2006-05-052-2/+47
|
* [189264] 6.11->6.21 upgrade, no db housekeeping threadsNoriko Hosoi2006-04-181-6/+4
| | | | | | | | The problem was check_db_version finds the existing DBVERSION is old and set dblayer_private->blayer_recovery_required = 1. If it's set, DBLAYER_CLEAN_RECOVER_MODE is processed properly. Unfortunately, it was "reinitialized" in the function read_metadata, which is called after check_db_version.
* 188931 - Use system Net-SNMP libraries on Linux systemsNathan Kinder2006-04-133-9/+34
|
* Bug(s) fixed: 186280Rich Megginson2006-04-1147-97/+137
| | | | | | | | | | | | | | | | | | | | | | | | | | Bug Description: ldapserver: Close potential security vulnerabilities in CGI code Reviewed by: Nathan, Noriko, and Pete (Thanks!) Fix Description: Clean up usage of sprintf, strcpy, fgets instead of gets, fixed buffer usage, etc., mostly in the CGI code and other user facing code (i.e. setup). Also, Steve Grubb told me about a GCC trick to force it to check printf style varargs functions, to check the format string against the argument string, for type mismatches, missing arguments, and too many arguments. In the CGI form argument parsing code, we needed to be more careful about checking for bad input - good input is supposed to look like this: name=value&name=value&..... &name=value. I don't think the original code was checking properly for something like name&name=value. There was another place where we were not checking to see if a buffer had enough room before appending a string to it. I had to change a couple of functions to allow passing in the size of the buffer. Fixed some issues raised by Noriko and Nathan. Platforms tested: RHEL4 Flag Day: no Doc impact: no QA impact: should be covered by regular nightly and manual testing New Tests integrated into TET: none
* Moved the SMARTHEAP macro setting from slapd local Makefile to global ↵Noriko Hosoi2006-04-071-27/+0
| | | | nsconfig.mk so that the makefile used for packaging gets the info.
* [186642] Directory Server Makefile updates for Internal build (Comment #6)Noriko Hosoi2006-04-071-8/+8
| | | | | | | | | | | | | 1) picked up the latest components including Cyrus SASL, Perldap, SetupUtil, and AdminServer 2) fixed the java paths for the internal build. 3) added an option -AA to HP-UX C++ compiler to adjust to SetupUtil build (use libstd 2) 4) replaced <iostream.h>, <strstream.h> and <fstream.h> with <iostream>, <strstream>, <fstream> respectively in ldap/cm/newinst/ux-dialog.cc 5) package svrcore and perldap, which were dropped in the current ldap/cm/Makefile 6) added a perl script genPerlDAPInf.pl to generate perldap.inf file
* [186642] Directory Server Makefile updates for Internal buildNoriko Hosoi2006-03-292-25/+29
| | | | | | | | | . removed the dependency on DBM . Updated components' RELDATEs. . Console jar file name adjustment: <brand>-<comp>-<dotversion>.jar . Fix for HP-UX IPF . Use non-branded SetupUtil; add resource file to change branding . Updated the code to call createSIE so that the right brand is passed to SetupUtil.
* 185811 - Missing const qualifiers were causing Windows compiler errorsNathan Kinder2006-03-201-4/+4
|
* 185811 - slapi_dn_find_parent needs to be added to libslapd.def for windows ↵Nathan Kinder2006-03-191-0/+1
| | | | build to compile
* 185811 - Don't check localuser config on WindowsNathan Kinder2006-03-191-0/+2
|
* 185811 - Need to exclude pwd.h include on windowsNathan Kinder2006-03-181-0/+2
|
* 185811 - Fixed windows build failures when compiling ↵Nathan Kinder2006-03-183-0/+6
| | | | slapd_chown_if_not_owner function
* [185477] ldif2db allows entries without a parent to be importedNoriko Hosoi2006-03-161-6/+6
| | | | if idl is NULL, changed add_op_attrs to always set IMPORT_ADD_OP_ATTRS_NO_PARENT in non-error case (err == 0 or err == DB_NOTFOUND)
* [185477] ldif2db allows entries without a parent to be importedNoriko Hosoi2006-03-151-3/+1
| | | | | | | The problem is that the new IDL code is not exposing the DB_NOTFOUND error when the index key is not found. Import checks if the result was DB_NOTFOUND when checking for the parent entry's entrydn index. This works fine with the old IDL code. (nkinder@redhat.com)
* 184585 - Adjusted the location of the connection lock when doing a sasl_rebindNathan Kinder2006-03-141-3/+3
|
* 184585 - SASL context needs to be disposed of and a new one created when ↵Nathan Kinder2006-03-142-0/+35
| | | | re-binding
generated by cgit (git 2.34.1) at 2024-11-19 00:16:58 +0000